The identification procedure assumes that the spot pattern of each individual is a unique distinguishing feature. Basically, the user points out the most distinguishing spots in each image. This spot pattern is stored in what is called a ‘fingerprint file’. In the identification step, the resulting spot pattern is matched with the spot patterns of all the known individuals in the database. To be able to correct for differences in viewing angle and distance, the user is also required to point out three reference points. Using a linear transformation it is now possible to compare two shark images in roughly the same coordinate system. A comparison of two individuals basically comes down to finding corresponding spot pairs in this coordinate system. From these pairs a distance metric is calculated to be able to rank each shark image in the database.

I³S Classic is based on the assumption that all spots are more or less the same size and close to circular. For Ragged Tooth Sharks and Whale Sharks this assumption works. However, other animals like Manta rays or the Ocean Sunfish, have a wide variety in spots concerning their shape and size. Our new tool, I³S Manta extends the use of I³S to these kinds of marine species by allowing the researchers to extract information about spot size, shape and direction and use this information in the comparison with a database.